Data Augmentation and Hyperparameter Tuning for Low-Resource MFA
By: Alessio Tosolini, Claire Bowern
Potential Business Impact:
Improves computer understanding of rare languages.
A continued issue for those working with computational tools and endangered and under-resourced languages is the lower accuracy of results for languages with smaller amounts of data. We attempt to ameliorate this issue by using data augmentation methods to increase corpus size, comparing augmentation to hyperparameter tuning for multilingual forced alignment. Unlike text augmentation methods, audio augmentation does not lead to substantially increased performance. Hyperparameter tuning, on the other hand, results in substantial improvement without (for this amount of data) infeasible additional training time. For languages with small to medium amounts of training data, this is a workable alternative to adapting models from high-resource languages.
Similar Papers
From Scarcity to Efficiency: Investigating the Effects of Data Augmentation on African Machine Translation
Computation and Language
Improves translation for African languages.
Fine Tuning Methods for Low-resource Languages
Computation and Language
Helps AI understand and use other languages better.
Frustratingly Easy Data Augmentation for Low-Resource ASR
Computation and Language
Makes talking computers understand rare languages better.